a pi using a micro

Sounds like a great idea, but be aware the posted TX hex code on his website for the TPIMD is faulty, you have to use the updated version.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,38170.msg235336.html#msg235336

I don't believe an Arduino would work. Too slow. They are an interpretive processor. You need a native code processor to get the speed required.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,38170.msg235362.html#msg235362

I do. Each instruction takes at least 4 clock cycles with some taking as many as 50. The Arduino is an interpretive processor using a BASIC interpreter.

Do try it though. I enjoy hearing about spinning wheels and brick wall scaling.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,38170.msg235417.html#msg235417

This is primarily a site for users of detectors not designers. There are only a handful here who would even be interested in assisting you but quite a few interested in seeing what you learn along the way. As to your theory that 4 processors would help you along, perhaps they would but you have to determine what each is going to do first. You can't just whack four processors on a board and expect it to work. That reminds me of the folks who come along here and want to know how to wire an LCD display into their current DIY project.

Start by designing a functional block diagram of your proposed machine. I have dozens of years experience in electronic design and embedded control but lack the time to help you so good luck.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,38170.msg235423.html#msg235423

Mate, get this in your head. This project is beyond the scope of this forum and I personally am not willing to dedicate any more time to such a project as you suggest. I have already been down that path and do not want to repeat my lack os success due to slow processor speeds. Unless you put a Pentium on the job, it's just making 4 machines talk to each other which, sad to say, takes up far too much processor time. No matter how fast your processors are, the more you add, the more they have to communicate with each other and when they do, they use more time. It just doesn't work as good as it sounds like it should.

But you go ahead and learn on your own instead of taking my word for it. Today, you have some advantages with multicore processors that have up to 8 processors inside a single chip. Their overhead is much better because internal bussing takes place but the tradeoff in cost is hardly worth it. A single chip costs over $1000 right now and I don't foresee them getting any less costly for two reasons. First is that they are horribly expensive to make and test. Second is the fact that they are really not all that useful so there is no queue of engineers wanting to use them. Until that happens, the price will stay high.

Good luck to you. Don't try to convince me that you have the latest and greatest idea. You'd be better served to read about the discussion here and see the conclusions that have resulted in those discussions. In other words, your idea has already been discussed here in many, many topic threads. It's not a new idea.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,38170.msg235447.html#msg235447